A realistic website project timeline starts with your calendar
A small business site takes six to ten weeks from signature to launch when everything goes normally, and normal means the client answers within two or three days and delivers text and photographs on the agreed dates. The building is rarely what takes the time.
Break a typical ten week project into where the hours sit and the picture is uncomfortable. Perhaps three weeks of design and code, perhaps a week of testing and fixes, and the rest spent waiting. Waiting for text. Waiting for a decision. Waiting for the photographs from the shoot that had to move because the workshop was busy.
That is not a complaint about clients, it is how the work is shaped. A builder can only lay out material that exists. Knowing this before you start is what turns an optimistic schedule into one you can actually keep.
The five stages of a normal project
Most projects move through the same stages whoever builds them. A short discovery phase where the goals, the structure and the audience are agreed. Content, where the text and the pictures are produced. Design, where the look and the page types are settled. Build, where it becomes a working site. And launch, which includes testing, the move to the real domain, and the week of small fixes afterwards.
The stages overlap in practice. Design usually starts before all the text is final, using the structure as a guide, and the build starts on the layouts that are already approved.
What must not overlap is approval. A design approved twice, once in week three and again in week seven after somebody senior finally looked, is the most expensive event in a website project. Get the right people in front of the work early. Our walkthrough of how a business website gets built covers what happens in each stage in more detail.
The parts that always take longer
Five things run over on almost every project, and none of them are code.
- Text. Always. Whoever writes it, the first draft arrives later than planned and the review takes two rounds more than anyone expected.
- Photography. The shoot depends on weather, on a tidy workshop and on people being available, and in Finland in November it also depends on the light.
- Approvals. One decision maker is fast. A board, a family business with three opinions, or an owner who is travelling adds a week per round.
- Third parties. Anything that involves another supplier moves at that supplier's speed, not yours.
- The last ten per cent. The small print pages, the error messages, the confirmation emails, the language switch on one awkward page.
Build the slack into the plan rather than pretending. A schedule with no slack is a schedule that is already late.
Why content is the real bottleneck
Because content is the only part of the project that cannot be bought with more hours from the builder. It has to come out of somebody's head, and that somebody is usually the busiest person in the company.
A plumbing firm in Vantaa can describe twenty years of work in a conversation. Writing four hundred words about it, at nine in the evening after a full day on site, is a different proposition. This is where projects sit still for three weeks with nobody doing anything wrong.
There are two ways out. Book two hours for an interview and let the writer turn the recording into pages, which is fast and produces better text than most owners write about themselves. Or set a hard date for your own draft and treat it like a customer deadline. What does not work is leaving it open, because open means never.
What you have to deliver, and when
Ask for this list in week one and put dates on it. A typical small business project needs the following from the client side.
- Logo files in the original format, and brand colours if they exist.
- Your business ID, registered name, address and the contact details you want public.
- Access to the domain, or the details of whoever holds it.
- Text for each page, or two hours for the interview that replaces it.
- Photographs, or a booked date for the shoot.
- Any legal text you already have, such as terms or a privacy policy.
- A named person who can approve, and a deputy for when that person is away.
The domain access line is the one that most often derails a launch, so start it in week one rather than the week you need it. We wrote about the traps in who owns your website and your domain.
Photography has its own calendar
Book the shoot in the first week of the project, not when the design is ready. Photographers in Finland are booked weeks ahead, and the good daylight in winter is a narrow window between mid morning and mid afternoon.
Think about what has to be in the pictures before the date. A clean van, a tidy bench, people in the clothes they actually wear, the product photographed against something plain. An hour of preparation the evening before is worth more than an extra hour of shooting.
A second language adds a round, not a week
Translation is quick. Approving translation is not. The Finnish version needs somebody who will actually read it and take responsibility for it, and finding that person late is what stretches multilingual projects.
Sequence it properly. Finalise the source language first, then translate, then build both. Translating text that is still changing means paying for it twice and finding three versions of the same sentence at launch.
Bookings, shops and integrations change the maths
Anything that connects to another system adds unpredictable time, because part of the schedule now belongs to somebody who has never heard of your project. Payment providers have onboarding checks. Accounting systems have their own support queues. A supplier product feed arrives in whatever format the supplier feels like.
Start those applications in week one. A payment provider that needs identity checks and a bank connection can take longer than the entire build, and it is a miserable reason for a shop to sit finished and unusable.
Testing also grows. A brochure site is checked by looking at it. A booking flow has to be tried on a phone, with a cancellation, over a public holiday, in both languages, with a full calendar and an empty one.
The domain and hosting handover can stall a launch
Sort out access before the last week. The classic Finnish version of this problem is a domain registered years ago by a nephew who has since emigrated, or held in the account of the previous supplier who is in no hurry to help.
Recovering control can take days of email and identity checks, and none of it is technically difficult. It is just slow, and it happens precisely when everyone wants to go live.
Check on day one whether you can log in and see the domain in your own name. If you cannot, that is the first task of the project, not the last.
A realistic schedule
For a six to ten page site for a small business in Finland, with one language and a shoot to organise, a schedule that holds looks roughly like this.
- Week 1. Kick off, structure agreed, content list issued with dates, photographer booked, domain access checked.
- Weeks 2 and 3. Text written or interviews recorded, shoot happens, first design of the home page and one inner page.
- Week 4. Design approved by the named approver. Remaining page types designed.
- Weeks 5 and 6. Build, with pages filled as final text arrives.
- Week 7. Testing on real phones, forms, speed, search basics, accessibility checks, corrections.
- Week 8. Launch, then a week of small fixes and measurement.
Add two to three weeks for a second language, two to four for a shop or a booking system, and a week for every approver beyond the first. Subtract nothing for enthusiasm.
What a rush job actually costs
Speed is buyable, and it is paid for in three currencies. Money, because overtime and reordered schedules cost more. Scope, because something gets cut. And quality, because the parts that get skipped are always the invisible ones, which means testing, speed and accessibility.
Sometimes rushing is right. A trade fair, a season that starts in April, a competitor about to open. When that is the case, cut scope deliberately rather than cutting care. Launch five excellent pages and add the rest in June.
What rarely works is keeping the full scope and compressing the calendar. That produces a launch followed by two months of fixing, and the fixing is visible to your customers.
The Finnish calendar is part of the plan
July is quiet, and planning a launch through it is planning around people who are not there. The weeks around Midsummer and the stretch from before Christmas into early January are similar, and the approvals you need will simply wait.
Use the quiet weeks for the work that does not need anyone else. Writing, photograph selection, structure. Keep the rounds that need decisions for weeks when the decision makers are at their desks.
Working backwards from a season helps. If your business peaks in spring, the site should launch in February so there is time to fix what the first real traffic reveals.
How to keep the project moving from your side
Three habits shorten a project more than any tool. Answer within two working days, even if the answer is that you need a week. Send all your comments on a round at once rather than in a trickle. And protect two hours a week in your own calendar for this project until it launches.
Ask your supplier for a single shared list of what is waiting on whom, and look at it weekly. Most delays are not decisions, they are questions nobody realised were open.
If you are planning a project and want a sanity check on the dates before you commit to them, tell us what you are working towards. A schedule that survives contact with a real calendar is worth more than an optimistic one.